Research On Fast Assessment Of Available Transmission Capacity Of Wind Power Grid Connected System

With the continuous development of renewable energy and the increasing sophistication of wind power technology in the world,wind power gradually occupies a place in the field of power generation,and countries around the world are scrambling to establish many wind farms of different scales.However,the grid connection of wind farms has brought many noticeable problems to the safe and stable operation of the original power grid,including but not limited to the randomness and discontinuity of wind farm output caused by natural uncontrollable factors of wind power.Therefore,it is necessary to study the calculation method of available transmission capacity(ATC)of large-scale wind farms,and then evaluate the calculation results,so as to provide an important reference for the operation and planning of power grid,so that it can operate safely and reliably.In this thesis,from the perspective of studying the fast calculation method of available transmission capacity,based on the impact of many uncertainties caused by wind power integration on the transmission capacity of the original power system,the following is the main research content and results of this thesis.Firstly,the probability density model of wind power is constructed.Due to the "tailing"phenomenon of wind power output,the commonly used single distribution is not very good for its fitting effect,so this paper uses the Gaussian mixture model,Then EM(expectation maximization)algorithm is used to solve the parameters of GMM model.Mae and RMSE are introduced to measure the fitting accuracy of GMM model.Secondly,in view of the time-consuming and low accuracy of available transmission capacity calculation,some literatures use Monte Carlo simulation method based on Markov chain(MCMC)to solve this problem.This thesis studies a ATC calculation method based on low rank approximation theory(LRA).This method performs deterministic ATC calculation based on optimal power flow method on the basis of a small number of system samples According to the statistical theory,the LRA Calculation Agent model is constructed,through which a large number of ATC samples are generated,and ATC related indicators are obtained.an example is given to verify that the proposed method is more efficient than MCMC method in solving ATC of wind power grid connected system.Finally,combined with LRA method,the available transmission capacity of grid connected wind power system with different capacity and location of wind farm is quickly evaluated.